Pitney Bowes
MapInfo Pro™ Monthly Journal
Something a bit more advanced: From Points to Regions (with Famous Mathematicians)

Peter This article is suitable for users of any version of MapInfo Pro.

Let's say you have set of point data for which you would like to estimate regional coverage for each point. Let's also say that a data set of regions is not available for you. This article will show how MapInfo Pro can build a set of contiguous regions from a set of points.

An example of where this might be helpful is if you have a postal set of data in the form of points but no regions are provided.

Not buffers...

One way of building regions out of points is to create buffers. However, buffers have a few downsides such as not necessarily covering the entire area of interest and creating overlaps. (See the screen shot below).


Famous Mathematicians

As an alternative to buffering you can create Voronoi polygons. Voronoi polygons represent the closest area to any one point. OK, that is not a strict and rigorous definition but I think serves our purposes here! Voronoi is the surname of one mathematician that came up with the formula to build polygons from points. Other mathematicians that explored this concept include Dirichlet and Thiessen. Wikipedia can tell you more, if you are interested.

The Quick Way (which may not be good enough)

Want to create Voronoi polygons quickly? Make a layer editable. (The cosmetic layer will do). Select some points.

In the 32-bit versions: On the Table menu, click on Voronoi.

In the 64-bit versions: On the SPATIAL tab, in the Regions drop-down, click on Voronoi.

That's all there is to it ... ... ...

But ... the Voronoi polygons are created for an area equal to the minimum bounding rectangle around your input dataset. Most likely this will not result in a coverage that includes your entire area of interest.


The More Thorough Way

It is possible to specify the area for which your Voronoi polygons should extend to. For our purposes here we will call this the area of interest.

Here is a description of the process
  1. Make the layer containing the area of interest Editable and select the region that makes up your area of interest. (Don't worry, you are not going to make any changes to this layer.)
  2. Select the region that represents your area of interest.

    One important point about this step is that you can only select a single region. As such you may need to take some extra steps to build the desired region first, if what you need is not available. For example you may need to Combine objects together.
  3. Next you need to set your selected region as the Target.

    In any version of MapInfo Pro you can do this with the Ctrl + T keyboard shortcut.

    If you are not a keyboard jockey the Set Target command is in the Objects menu in the 32 bit version. In the 64 bit versions it is on the SPATIAL tab.

    Now we are ready to create the Voronoi polygons.
  4. Now, create the Voronoi regions.

    In the 32-bit versions: On the Table menu, click on Voronoi.

    In the 64-bit versions: On the SPATIAL tab, in the Regions drop-down, click on Voronoi.
  5. Choose the input table (the table containing the point dataset) and your desired output table. Quite often, the output table will be a new table. You can create a new table during this process.
  6. Specify how to display this new table (a common setting here is to add the new table to the Current Map window). Also make sure to copy the table structure from your input table. This allows you to copy all of the attribute data from the input point table into the new Voronoi polygon table.
  7. The Table Structure dialog box appears. You can alter the structure if needed.
  8. Name the new table and choose where to store it.
  9. Specify how to move the data from your input table to the new Voronoi table. If you choose to copy the structure to the new table, the data will, by default be copied between the two tables.
  10. MapInfo Pro will calculate the new Voronoi polygons and display them as you chose earlier.
In the first example, the result was a set of Voronoi polygons that represented the extent of the input point dataset. In contrast, the Voronoi polygons below extend to a chosen boundary.


Join our LinkedIn Group

Have a question? Want to request a topic to be covered in this journal? Come on over to "The MapInfo Pro Journal LinkedIn Group" and let us know.

Article by Peter Horsboll Moeller, EMEA Channel Enablement Specialist
When not writing articles for "The MapInfo Pro" journal, Peter helps Pitney Bowes Software customers to get the most from their software and is a prolific contributor to the mapinfo-l Google Group. When not working he is an aficionado of fine whiskey and good cigars.